Research On Low Altitude Photography Technology In Mountain Highway Disaster Survey

The only mountain roads are often way to contact the outside world in remote areas.Disasters cause road collapse and traffic disruption,which seriously affects the lives of mountain people.Therefore,the investigation of disasters on mountain roads is particularly important.This paper analyzes the types of mountain highway disasters in high-resolution images,and describes the characteristics of the disaster features in combination with the features of the features.The low-altitude photogrammetry technology is used to obtain the digital elevation model and digital orthophotos model of the disaster area.The Classification based on fuzzy rules obtains the shape data of the roads and landslides in the study area,and superimposes the extracted disaster area vector data with the digital elevation model to calculate the disaster area and earthwork volume of the landslide body.The research content and results mainly include the following aspects:(1)In view of the emergency needs of mountain road disaster investigation,this paper uses UAV to obtain high-resolution image data,and uses POS automatic three-solution method to reconstruct the 3D ground model without ground control points,and check through field collection.The point is used to evaluate the accuracy of the model and obtain the thematic data of the study area.The experiment shows that the generated data is of good quality.(2)In view of the lack of ground information caused by vegetation cover during the extraction process of mountain road disaster area,and the serious damage of geometric features,this paper uses a new method based on point cloud model to remove vegetation covers the covered vegetation in digital positive photography,and solves the problem that the mountain roads in vegetation coverage are difficult to extract.Through the comparative analysis of the disaster extraction results,the method can present the real road information under the vegetation cover,and provide a new theoretical reference for the disaster investigation of mountain roads.(3)For the uncertainty of landslide characteristics,this paper uses the object-oriented classification method based on the feature description of morphological features to extract the landslide area in the study area by constructing the fuzzy feature rule set of landslide body,using visual interpretation and The method of field test evaluates the extraction results.Compared with the pixel-based classification method,the method eliminates the “salt and salt phenomenon”.Compared with other object-oriented methods,the classification method can better reflect the feature of the feature.(4)The digital elevation model generated based on low-altitude photography technology has higher precision than platforms such as satellite remote sensing and aerial photography.The disaster vector data acquired by the image classification method is superimposed with the digital elevation model to calculate the disaster area and volume of the disaster body,and provide parameter indicators for determining the disaster level.The research results show that low-altitude photography technology can meet the emergency needs of mountain road disasters,which can reduce the labor cost compared with the field measurement.Compared with the remote sensing information extraction method,it can basically solve the problem that the vegetation information is difficult to extract due to vegetation coverage.This technology has certain research value in the investigation of mountain road disasters.